Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

OPEN LETTER

Insights from quantitative analysis and mathematical modelling on the proposed WHO 2030 goals for soil- transmitted helminths [version 2; peer review: 2 approved]

NTD Modelling Consortium discussion group on soil-transmitted helminths

v2 First published: 23 Oct 2019, 3:1632 Open Peer Review https://doi.org/10.12688/gatesopenres.13077.1 Latest published: 04 Dec 2019, 3:1632 https://doi.org/10.12688/gatesopenres.13077.2 Reviewer Status

Invited Reviewers Abstract Soil-transmitted helminths (STHs) are a group of parasitic worms that 1 2 infect humans, causing a wide spectrum of disease, notably anaemia, growth retardation, and delayed cognitive development. The three version 2 main STHs are , and hookworm ( (revision) and ). Approximately 1.5 04 Dec 2019 billion people are infected with STHs worldwide. The World Health Organization goal for 2030 is morbidity control, defined as reaching version 1 <2% prevalence of medium-to-high intensity infections in preschool- 23 Oct 2019 report report age children and school-age children (SAC). Treatment guidelines for achieving this goal have been recommended. The Neglected Tropical 1. Rubina Imtiaz , The Task Force for Global Diseases Modelling Consortium has developed mathematical and statistical models to quantify, predict, and evaluate the impact of Health, Decatur, USA control measures on STHs. These models show that the morbidity target can be achieved following current guidelines in moderate 2. Suzy J. Campbell , Evidence Action: prevalence settings (20-50% in SAC). In high prevalence settings, semi- Deworm the World Initiative, Washington, annual preventive chemotherapy (PC) ideally including adults, or at USA least women of reproductive age, is required. For T. trichiura, dual therapy with albendazole and ivermectin is required. In general, Any reports and responses or comments on the stopping PC is not possible without infection resurgence, unless article can be found at the end of the article. effective measures for improved access to water, hygiene, and sanitation have been implemented, or elimination of transmission has been achieved. Current diagnostic methods are based on egg counts in stool samples, but these are known to have poor sensitivity at low prevalence levels. A target threshold for novel, more sensitive diagnostics should be defined relative to currently preferred diagnostics (Kato-Katz). Our analyses identify the extent of systematic non-access to treatment and the individual patterns of compliance over multiple rounds of treatment as the biggest unknowns and the main impediment to reaching the target. Moreover, the link between morbidity and infection intensity has not been fully elucidated. By providing more insights on all the above, we aim to inform discussions on the goals and treatment guidelines for STHs.

Page 1 of 13 Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

Keywords Soil-transmitted helminths, WHO guidelines, morbidity control, NTD Modelling

This article is included in the 2030 goals for neglected tropical diseases collection.

Corresponding author: NTD Modelling Consortium discussion group on soil-transmitted helminths ([email protected]) Competing interests: No competing interests were disclosed. Grant information: This work was supported by the Bill and Melinda Gates Foundation [OPP1184344]. Luc E. Coffeng was supported by a personal grant from the Dutch Research Council (Nederlandse Organisatie voor Wetenschappelijk Onderzoek [016.Veni.178.023]). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Copyright: © 2019 NTD Modelling Consortium discussion group on soil-transmitted helminths. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. How to cite this article: NTD Modelling Consortium discussion group on soil-transmitted helminths. Insights from quantitative analysis and mathematical modelling on the proposed WHO 2030 goals for soil-transmitted helminths [version 2; peer review: 2 approved] Gates Open Research 2019, 3:1632 https://doi.org/10.12688/gatesopenres.13077.2 First published: 23 Oct 2019, 3:1632 https://doi.org/10.12688/gatesopenres.13077.1

Page 2 of 13 Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

The three main STHs are roundworm (Ascaris lumbricoides), REVISED Amendments from Version 1 whipworm (Trichuris trichiura) and hookworm (Necator Minor edits were made to clarify the following issues highlighted americanus and Ancylostoma duodenale). STHs reduce the by Reviewer 2: nutritional status of infected individuals1. In particular, infected 1) Achieving 75% community-wide treatment is especially children can be affected by reduced physical fitness and important to reach the morbidity target for hookworm. impaired growth and cognitive development1. Hookworm 2) Elimination of transmission can be reached in low-transmission infection in women of reproductive age (WRA) can lead to settings. The study on which this insight is based used the basic severe anaemia1. Infections with A. lumbricoides and hookworms reproductive number as a measure of transmission intensity, not prevalence. can be treated effectively with benzimidazole drugs (albenda- 3) All simulation studies have been run for the different STH zole, mebendazole). However, benzimidazoles are less effec- species (Ascaris lumbricoides, Trichuris trichiura, and hookworm tive against T. trichiura. Dual treatment with albendazole and species). Insights are, therefore, species-specific. ivermectin increases treatment efficacy for T. trichiura2–4. 4) The results and conclusion with regard to the importance of Currently, albendazole and mebendazole are donated to the WASH are based on simulations for the impact of sanitation and World Health Organization (WHO) for distribution to affected hygiene measures in isolation and in combination. populations. Any further responses from the reviewers can be found at the end of the article The WHO has announced morbidity control as the main public health target for STHs to be achieved by 2030. According to the most recent WHO guidelines, morbidity control is Disclaimer defined as <2% prevalence of medium-to-high intensity (M&HI) The views and opinions expressed in this article are those of the infections in preschool-age children (preSAC) and school-age authors and do not necessarily reflect the views of the World children (SAC). WHO treatment guidelines advise preventive Health Organization. Publication in Gates Open Research does chemotherapy (PC) by mass drug administration (MDA) to not imply endorsement by the Gates Foundation. achieve morbidity control. Previously, WHO recommended school-based PC without including adults. The most recent Background guidelines recommend PC targeted at preSAC, SAC and Soil-transmitted helminth (STH) infections are caused by WRA. The frequency of PC is based on the prevalence of several species of parasitic worms that are transmitted by eggs STH infections in SAC prior to the start of treatment (see present in human faeces, which contaminate the soil in areas decision tree in Figure 1 for WHO guidelines up until 2019). with poor sanitation. STHs cause some of the most common The recommended PC coverage is 75% in all targeted infections, with about 1.5 billion people infected worldwide1. populations.

Figure 1. World Health Organization (WHO) decision tree showing the current WHO guidelines to achieve morbidity control in school-aged children (SAC) using 75% coverage in SAC (black and orange boxes). The bright red boxes represent the modified guidelines assuming 75% community-wide coverage (pre-SAC, SAC, and adults to replace the middle row of boxes that represent the current guidelines. This figure has been adapted fromFigure 1 in 5 under a CC-BY 4.0 license.

Page 3 of 13 Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

Mathematical models of STH transmission dynamics and achieve the 2020 morbidity target in settings where the the impact of interventions have been developed to evaluate prevalence was moderate (20% to 50% in SAC) prior to the start optimal treatment strategies for achieving the WHO goals. The of PC5. For higher prevalence settings, community-wide PC Neglected Tropical Diseases Modelling Consortium (NTDmc) and/or targeting of WRA will be necessary to achieve the funded by the Bill and Melinda Gates Foundation brings morbidity target, especially for hookworm5,8, and/or dual together research groups from multiple scientific institutions therapy with albendazole and ivermectin for T. trichiura5,9. working on neglected tropical diseases (NTDs), including STHs. Implementing PC twice-yearly also increases chance to achieve Modelling groups based at Erasmus Medical Center (EMC) the morbidity target for STH5. Scaling down or stopping PC in Rotterdam and Imperial College London (ICL) have led the as per WHO treatment guidelines is very likely to lead to resur- recent work on STHs. A model comparison was carried out gence of infection to levels above the morbidity target, unless for the EMC and ICL STH models6. Moreover, joint papers transmission conditions are addressed with water, sanitation evaluating WHO treatment guidelines and monitoring and and hygiene (WASH)10 or elimination of transmission (EOT) evaluation strategies have been published5,7. Predictions have is achieved5. If this is not feasible, PC needs to be sustained10. been made for each STH species to account for inter-species Impact assessments potentially need to be repeated at regular variability. In these predictions it is important to note that intervals. Accurate measurements of access and compliance models predict true prevalence while surveillance data make to PC remain essential to evaluate and sustain achievement of predictions that are sensitive to the diagnostic method employed. the targets. It is also important to note the poor sensitivity of The predictions of both models are largely comparable, although Kato-Katz at low prevalence (models predict true prevalence). the EMC model is more optimistic about the additional impact of community-wide vs. targeted (pre-SAC, SAC and WRA) Practical implications of the proposed goals deworming, which can be readily explained by differences in Measuring the target assumptions about how rates at which individuals contaminate Geospatial analyses of recent large-scale epidemiological the environment vary with age. studies of STH prevalence11 show that prevalence heterogeneity is considerable within PC implementation units. Our simulations The WHO has proposed new goals for NTDs, including new suggest that to evaluate PC impact, a sufficient number of villages control targets for STHs in the year 2030. Using insights from should be sampled in each implementation unit7 for an accurate recent modelling work we discuss the feasibility of reaching the assessment of the prevalence (number of villages depending on morbidity target following current guidelines and the risks that geospatial variation). need to be mitigated to maintain the target (summary in Table 1). The indicator for the morbidity target will be measured with Insights from modelling: Lessons from the past 10 Kato-Katz (or any other validated quantitative technique). years for the next 10 years However, the number of slides/samples used strongly affects the Our modeling and epidemiological data analyses have shown measured prevalence12,13. PCR methods, although expensive at that the current WHO treatment guidelines are sufficient to present, are a much more sensitive diagnostic at low prevalence.

Table 1. Summary of modelling insights and challenges for reaching the WHO 2030 goal for soil-transmitted helminths.

Current WHO Goal (2020) Morbidity control: <1% prevalence of M&HI infections in preSAC and SAC Proposed new WHO Goal Morbidity control: <2% prevalence of M&HI infections in preSAC and SAC (2030) Is the new target technically Yes, in moderate prevalence settings (20–50% in SAC) in the absence of systematic non-access to feasible under the current treatment. For highly endemic settings (prevalence ≥50%), semi-annual PC, including adults or at least disease strategy? WRA (hookworm), and/or dual PC (Trichuris) are required. A lot depends on the diagnostic used and these targets may have to be revised if the employment of PCR methods reveals much higher levels of infection. Are current tools able to Yes, although there is a need to test and identify the optimal design for surveys based on Kato-Katz reliably measure the target? (how many stool samples per person, how many slides per sample) and PCR for the specific purpose of evaluating the target and intermediate markers of progress (which are based on prevalence of any infection instead of M&HI). What are the biggest Levels of systematic non-access or non-compliance to treatment and its impact on achievement of the unknowns? target; the link between morbidity and present and past cumulated infection intensity and how the current parasitological target translates to actual morbidity levels; epidemiological situation in settings with pre- control prevalences <20% (meaning no implementation of PC) as PC in other areas continues. What are the biggest risks? Systematic non-access and non-compliance to treatment, low coverage and resurgence after reducing treatment frequency, lack of community-wide treatment, especially when hookworm is the dominant infection if the intrinsic transmission potential is high.

WHO, World Health Organization; M&HI, medium-to-high intensity; preSAC, preschool-age children; SAC, school-age children; PC, preventative chemotherapy; WRA, women of reproductive age; PCR, polymerase chain reaction.

Page 4 of 13 Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

The indicator threshold would be more meaningful if linked to with low systematic non-access/non-compliance to PC and little a standardized diagnostic procedure, or to the true prevalence of coverage heterogeneity within PC implementation units. infection from which thresholds for specific diagnostic procedures Modelling suggests that the timeline for achieving the target and sampling designs can be derived. is expected to be longer if there is re-importation of disease, e.g. by migration for areas with low or no treatment Current egg counting methods suffer from considerable coverage16. Meeting the target may require coordination of measurement error, which is compounded by high variation in national STH programmes at country borders due to human egg density between and within persons over time13–15, meaning movement. that an observed prevalence of M&HI can be well above the 2% target by chance. This is further compounded by an increase in Ability to sustain achievement of the goal inter-individual variation in egg counts as infection prevalence After stopping or scaling down treatment (which is an option in goes down during PC (likely due to systematic non-access to the current WHO guidelines, see Figure 1), infection levels are PC). Further, modelling suggests that “prevalence of any infection likely to bounce back within one to two years17. Thus, it may in preSAC and SAC” combined with a higher target threshold is not be possible to decrease the number of required tablets as a more informative indicator (higher positive predictive value) proposed as a new WHO target5,8. This is further complicated for meeting the morbidity target and would require a smaller by population growth between now and 2030, which could sample size because of a higher statistical power7. necessitate a further increase in the number of treatments required for pre-SAC and SAC. See Figure 1 for an alternative Timeline to achieve the target decision-tree based on recent modelling. The 2030 morbidity target will be achievable in some countries. The frequency and duration of PC and implied resources Our analyses suggest that uptake of effective WASH is needed depend on baseline prevalence and achieved coverage plus to sustain the gains made by PC in the longer term10. If EOT is patterns of individual compliance to treatment5. not achieved and PC is stopped or scaled down in the absence of effective WASH, the probability of resurgence is very high5. Technical feasibility In the absence of effective WASH interventions, the sustainability Treatment guidelines will lead to the achievement of the target of the morbidity targets is undermined by human population in some communities, but not in all5. Current WHO guidelines movement unless PC is continued indefinitely16. do not call for treatment in low prevalence settings (<20%). However, these areas may still have a prevalence of M&HI Considerations of cost >2% in preSAC, SAC and WRA14. In addition, epidemiological Child-targeted treatment for hookworm is cost-effective at data from the Tumikia study suggest that with lower prevalence, reducing morbidity in children, even in high-transmission the prevalence of M&HI is relatively higher due to increasing settings18,19. Community-wide treatment is predicted to be more aggregation of parasites as MDA coverage rises, likely due to cost-effective in the longer term with respect to the overall a small proportion of persistent non-compliers to treatment. A morbidity case-years prevented than child-targeted treatment, revision of the 20% threshold downwards seems desirable. as the main hookworm disease burden lies in adults18,19. Annual co-administration of albendazole/mebendazole with ivermectin WHO guidelines for moderate-prevalence settings suggest is predicted to be more cost-effective than semi-annual annual PC of young children, preSAC, SAC and WRA. This albendazole/mebendazole treatment for reducing the prevalence may be sufficient to reach the morbidity target for settings where of heavy T. trichiura infections in SAC9. In general, achieving coverage is sufficiently high (75%) in the absence of systematic high coverage and good individual compliance in annual treat- non-compliance to treatment. ment rounds may be more cost-effective than treating twice a year with lower coverage. For high-prevalence settings (>50%), WHO guidelines suggest semiannual treatment. Here the morbidity target is less likely Risks that need to be mitigated to achieve the stated to be achieved following current guidelines, especially for goals hookworm and T. trichiura. As the main burden of hookworm Population movement can re-import infection into a geo- infection lies in adults, the morbidity targets will only be graphical area that has previously reached morbidity control or reached when also treating adults as a whole5, not just WRA8. EOT. Measures to mitigate this risk include aiming for evenly Control of T. trichiura will require community-wide treatment high coverage across implementation units and co-ordination of with albendazole3 or dual treatment with ivermectin and a programmes across country borders16. benzimidazole3,5. If systematic non-access to treatment remains high, meeting the target may not be feasible8. Systematic non-access and non-compliance to treatment in repeated rounds of MDA and predisposition to heavy infection Operational feasibility will create a pool of individuals with high infection burden that Reaching the milestone in 2030 will require community-wide can re-infect others8,10,20–24. Increasing access as well as coverage coverage and/or targeting of WRA (especially for hookworm), will be important for achieving the 2030 targets22,23.

Page 5 of 13 Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

Reducing frequency of treatment, as proposed in WHO Future work guidelines25 and the new goal for 2030 of reducing the number Future work that the NTDmc can contribute in support of the of tablets required for treating STHs, in the absence of EOT design and achievement of the WHO 2030 goals will focus and/or effective WASH measures (including both measures on: 1) an analysis of the value of different diagnostic methods reducing contamination and measures reducing exposure to and sampling strategies on M&E of STH morbidity targets and infection) can lead to rapid resurgence of infection prevalences predicting the probability of EOT; 2) understanding the role of to pre-treatment levels5,10,26–28. See Figure 1 for an alternative spatial heterogeneity in prevalence and coverage and human decision-tree based on recent model-based analyses. population movement on STH control programmes; 3) investigat- ing the risk of emergence of drug resistance as well as whether Discussion and how monitoring of drug efficacy may help, 4) quantifying The morbidity target is defined in terms of the prevalence of the link between infection intensity and morbidity; and M&HI. However, infection intensity does not necessarily reflect 5) assessing the importance of different patterns of individual morbidity accurately, as light infections can be associated compliance to treatment to achieving the WHO targets as data with non-negligible morbidity and the severity of symptoms becomes available from large-scale epidemiological studies associated with M&HI is highly variable29. Furthermore, and trials. Other proposed topics for future work include the current diagnostic tools have poor sensitivity at low prevalence impact of discontinuation of lymphatic filariasis programmes on levels. Defining targets depending on the diagnostic used seems STH, infection models encapsulating molecular epidemiology desirable in future policy formulation. data of who infects whom, defining threshold values for when systematic non-access and non-compliance causes failure to Transmission dynamic models with parameters estimated from achieve WHO targets, and development of transmission models cross-sectional and longitudinal epidemiological data show for Strongyloides stercoralis. that technically EOT is feasible for STH in some settings. It is predicted that EOT can be achieved in low-transmission Data availability settings where A. lumbricoides or T. trichiura are the No data are associated with this article. dominant parasites by annual treatment of SAC, assuming 80% effective coverage and random compliance at each round of treatment30. Where EOT is feasible, it may be more cost- Acknowledgements effective than continuous morbidity control, provided no NTD Modelling Consortium discussion group on soil-transmitted re-importation occurs19. In high transmission settings, helminths: community-wide treatment is predicted to be more effective Carolin Vegvari2, Federica Giardina1, Veronica Malizia1, (especially for hookworm) and more cost-effective. Sake J. de Vlas1, Roy M. Anderson2, Luc E. Coffeng1

Another new WHO goal for 2030 is control of strongyloidia- 1. Department of Public Health, Erasmus Medical Center, sis. This requires ivermectin treatment, which would particularly Rotterdam, The Netherlands benefit areas with high prevalence of T. trichiura. Currently, 2. London Centre for Neglected Tropical Disease Research, for policy assessments there is only epidemiological data on Department of Infectious Disease Epidemiology, Imperial College 31,32 Strongyloides stercoralis but no model-based predictions. London, London, UK As for the other STH, models will provide useful insights for policy formulation. Corresponding author: Luc E. Coffeng ([email protected])

References

1. Soil-transmitted helminth infections. [Internet]. [cited 2019 Jul 8]. Current and Modified World Health Organization Guidelines for the Control of Reference Source Soil-Transmitted Helminth Infections. Clin Infect Dis. 2018; 66(suppl_4): S253–S259. 2. Belizario VY, Amarillo ME, de Leon WU, et al.: A comparison of the efficacy of PubMed Abstract | Publisher Full Text | Free Full Text single doses of albendazole, ivermectin, and diethylcarbamazine alone or in 6. Coffeng LE, Truscott JE, Farrell SH, et al.: Comparison and validation of two combinations against Ascaris and Trichuris spp. Bull World Health Organ. 2003; mathematical models for the impact of mass drug administration on Ascaris 81(1): 35–42. lumbricoides and hookworm infection. Epidemics. 2017; 18: 38–47. PubMed Abstract | Free Full Text PubMed Abstract | Publisher Full Text | Free Full Text 3. Knopp S, Mohammed KA, Speich B, et al.: Albendazole and mebendazole 7. Giardina F, Coffeng LE, Farrell SH, et al.: Sampling strategies for monitoring and administered alone or in combination with ivermectin against Trichuris evaluation of morbidity targets for soil-transmitted helminths. PLoS Negl Trop trichiura: a randomized controlled trial. Clin Infect Dis. 2010; 51(12): 1420–8. Dis. 2019; 13(6): e0007514. PubMed Abstract | Publisher Full Text PubMed Abstract | Publisher Full Text | Free Full Text 4. Clarke NE, Doi SAR, Wangdi K, et al.: Efficacy of Anthelminthic Drugs and Drug 8. Coffeng LE, Bakker R, Montresor A, et al.: Feasibility of controlling hookworm Combinations Against Soil-transmitted Helminths: A Systematic Review and infection through preventive chemotherapy: a simulation study using the Network Meta-analysis. Clin Infect Dis. 2019; 68(1): 96–105. individual-based WORMSIM modelling framework. Parasit Vectors. 2015; 8(1): 541. PubMed Abstract | Publisher Full Text PubMed Abstract | Publisher Full Text | Free Full Text 5. Farrell SH, Coffeng LE, Truscott JE, et al.: Investigating the Effectiveness of 9. Turner HC, Truscott JE, Bettis AA, et al.: Analysis of the population-level impact

Page 6 of 13 Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

of co-administering ivermectin with albendazole or mebendazole for the 2017; 358: j4307. control and elimination of Trichuris trichiura. Parasite Epidemiol Control. 2016; PubMed Abstract | Publisher Full Text | Free Full Text 1(2): 177–187. 21. Dunn JC, Bettis AA, Wyine NY, et al.: Soil-transmitted helminth reinfection four PubMed Abstract | Publisher Full Text | Free Full Text and six months after mass drug administration: results from the delta region 10. Coffeng LE, Vaz Nery S, Gray DJ, et al.: Predicted short and long-term impact of Myanmar. PLoS Negl Trop Dis. 2019; 13(2): e0006591. of deworming and water, hygiene, and sanitation on transmission of soil- PubMed Abstract | Publisher Full Text | Free Full Text transmitted helminths. PLoS Negl Trop Dis. 2018; 12(12): e0006758. 22. Dyson L, Stolk WA, Farrell SH, et al.: Measuring and modelling the effects of PubMed Abstract | Publisher Full Text | Free Full Text systematic non-adherence to mass drug administration. Epidemics. 2017; 18: 11. Pullan RL, Smith JL, Jasrasaria R, et al.: Global numbers of infection and 56–66. disease burden of soil transmitted helminth infections in 2010. Parasit Vectors. PubMed Abstract | Publisher Full Text | Free Full Text 2014; 7: 37. 23. Farrell SH, Truscott JE, Anderson RM: The importance of patient compliance PubMed Abstract | Publisher Full Text | Free Full Text in repeated rounds of mass drug administration (MDA) for the elimination of 12. Levecke B, Brooker SJ, Knopp S, et al.: Effect of sampling and diagnostic effort intestinal helminth transmission. Parasit Vectors. 2017; 10(1): 291. on the assessment of schistosomiasis and soil-transmitted PubMed Abstract | Publisher Full Text | Free Full Text and drug efficacy: a meta-analysis of six drug efficacy trials and one 24. Wright JE, Werkman M, Dunn JC, et al.: Current epidemiological evidence for epidemiological survey. Parasitology. 2014; 141(14): 1826–40. predisposition to high or low intensity human helminth infection: a systematic PubMed Abstract Publisher Full Text | review. Parasit Vectors. 2018; 11(1): 65. 13. Knopp S, Mgeni AF, Khamis IS, et al.: Diagnosis of soil-transmitted helminths PubMed Abstract | Publisher Full Text | Free Full Text in the era of preventive chemotherapy: effect of multiple stool sampling and 25. Montresor A, World Health Organization, editors: Helminth control in school-age use of different diagnostic techniques. PLoS Negl Trop Dis. 2008; 2(11): e331. children: a guide for managers of control programmes. 2nd ed. Geneva: World PubMed Abstract Publisher Full Text Free Full Text | | Health Organization; 2011; 76. 14. Anderson RM, Schad GA: Hookworm burdens and faecal egg counts: an Reference Source analysis of the biological basis of variation. Trans R Soc Trop Med Hyg. 1985; 79(6): 812–25. 26. Anderson RM, May RM: Helminth infections of humans: mathematical models, PubMed Abstract Publisher Full Text population dynamics, and control. Adv Parasitol. 1985; 24: 1–101. | PubMed Abstract Publisher Full Text 15. Easton AV, Oliveira RG, Walker M, et al.: Sources of variability in the | measurement of Ascaris lumbricoides infection intensity by Kato-Katz and 27. Anderson R, Hollingsworth TD, Truscott J, et al.: Optimisation of mass qPCR. Parasit Vectors. 2017; 10(1): 256. chemotherapy to control soil-transmitted helminth infection. Lancet. 2012; PubMed Abstract Publisher Full Text Free Full Text 379(9813): 289–90. | | PubMed Abstract Publisher Full Text 16. Vegvari C, Truscott JE, Kura K, et al.: Human population movement can | impede the elimination of soil-transmitted helminth transmission in regions 28. Anderson RM, Turner HC, Truscott JE, et al.: Should the Goal for the Treatment with heterogeneity in mass drug administration coverage and transmission of Soil Transmitted Helminth (STH) Infections Be Changed from Morbidity potential between villages: a metapopulation analysis. Parasit Vectors. 2019; Control in Children to Community-Wide Transmission Elimination? PLoS Negl 12(1): 438. Trop Dis. 2015; 9(8): e0003897. PubMed Abstract | Publisher Full Text | Free Full Text PubMed Abstract | Publisher Full Text | Free Full Text 17. Thein-Hlaing, Than-Saw, Htay-Htay-Aye, et al.: Epidemiology and transmission 29. Campbell SJ, Nery SV, Doi SA, et al.: Complexities and Perplexities: A Critical dynamics of Ascaris lumbricoides in Okpo village, rural Burma. Trans R Soc Appraisal of the Evidence for Soil-Transmitted Helminth Infection-Related Trop Med Hyg. 1984; 78(4): 497–504. Morbidity. PLoS Negl Trop Dis. 2016; 10(5): e0004566. PubMed Abstract | Publisher Full Text PubMed Abstract | Publisher Full Text | Free Full Text 18. Turner HC, Truscott JE, Fleming FM, et al.: Cost-effectiveness of scaling up 30. Truscott JE, Hollingsworth TD, Brooker SJ, et al.: Can chemotherapy alone mass drug administration for the control of soil-transmitted helminths: a eliminate the transmission of soil transmitted helminths? Parasit Vectors. 2014; comparison of cost function and constant costs analyses. Lancet Infect Dis. 7: 266. 2016; 16(7): 838–846. PubMed Abstract | Publisher Full Text | Free Full Text PubMed Abstract | Publisher Full Text 31. Forrer A, Khieu V, Schär F, et al.: Strongyloides stercoralis is associated with 19. Turner HC, Truscott JE, Bettis AA, et al.: An economic evaluation of expanding significant morbidity in rural Cambodia, including stunting in children. PLoS hookworm control strategies to target the whole community. Parasit Vectors. Negl Trop Dis. 2017; 11(10): e0005685. 2015; 8: 570. PubMed Abstract | Publisher Full Text | Free Full Text PubMed Abstract | Publisher Full Text | Free Full Text 32. Schär F, Trostdorf U, Giardina F, et al.: Strongyloides stercoralis: Global 20. Moser W, Schindler C, Keiser J: Efficacy of recommended drugs against soil Distribution and Risk Factors. PLoS Negl Trop Dis. 2013; 7(7): e2288. transmitted helminths: systematic review and network meta-analysis. BMJ. PubMed Abstract | Publisher Full Text | Free Full Text

Page 7 of 13 Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

Open Peer Review

Current Peer Review Status:

Version 1

Reviewer Report 18 November 2019 https://doi.org/10.21956/gatesopenres.14216.r28134

© 2019 Campbell S. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Suzy J. Campbell Evidence Action: Deworm the World Initiative, Washington, DC, USA

The proposed World Health Organization (WHO) 2030 Roadmap goal of morbidity control for soil- transmitted helminths (STH) is defined as reaching <2% prevalence of moderate-to-heavy intensity infections (MHII) in preschool (PSAC) and school age children (SAC). In this manuscript, the NTD Modelling Consortium discussion group on STH provide a summary of insights from mathematical and statistical models into whether this goal is feasible to achieve, and to evaluate the optimal treatment strategies to reach it. In summary, based on two separate but comparable predictive models, the authors determine that the 2030 morbidity target could be achieved using current treatment guidelines in moderate settings (being those with 20-50% prevalence in SAC). Higher prevalence settings are deemed likely to require semi-annual preventive chemotherapy, ideally by including adults (or at least women of reproductive age, WRA), and combined albendazole- ivermectin against Trichuris trichiura. To prevent resurgence, preventive chemotherapy needs to be sustained, unless there are effective measures to improve access to water, sanitation and hygiene (WASH), or unless elimination of transmission (EOT) has been achieved. A target threshold for novel, sensitive diagnostic tests is required. The biggest unknowns, and greatest impediments to the target, are deemed to be systematic non-access to treatment and individual compliance over multiple rounds of mass drug administration (MDA). The authors recommend changes to the WHO decision trees, assuming 75% community-wide coverage.

Gaining this level of insight from the NTD Modelling Consortium is very useful and the article itself is well constructed. However, by modelling what would be required to reach the proposed 2030 goal, the authors highlight some key anomalies between the 2030 goal and the current international landscape. The first is that although WRA are not explicitly included in the goal definition, actions to achieve the goal for PSAC and SAC will require treatment of WRA, if not all adults, in certain high prevalence settings. There remains insufficient evidence at this time, particularly on cost-effectiveness, to scale up treatment of the WRA cohort via MDA campaigns. Further, in Figure 1, the authors suggest modified guidelines that assume 75% community-wide coverage. Achieving and sustaining 75% full community coverage would be very hard, and this is neither seen as particularly scalable nor necessary for a morbidity-control goal. WHO guidance is recommended to at-risk (targeted) populations and they stop short of recommending full

Page 8 of 13 Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

community treatments. As is known, drug donations and/or implementation advice to reach expanded cohorts are not provided, and examples of programmatic expansion are rare. Is the authors’ recommendation of 75% community-wide coverage applicable to all settings, or has this extended beyond what is required for the current morbidity control goal?

The authors are kindly requested to elucidate further on the statement (Discussion) that “it is predicted that EOT can be achieved in low-transmission settings where Ascaris lumbricoides or T. trichiura are the dominant parasites by annual treatment of SAC”. If this is the case, but this manuscript has investigated achieving the goal for scenarios above 20%, it seems unlikely that EOT would enter into consideration and thus, preventive chemotherapy would have to continue. The STH morbidity control goal is to eliminate heavy burden of disease, not eliminate a parasite. Whilst, biologically, EOT is feasible, this is not yet considered the best use of sparse resources, and it is an investment with a number of unknown factors that most countries would be inadequately prepared for.

If neither WASH access nor EOT are feasible, then preventive chemotherapy needs to be sustained. This recommendation for indefinite preventive chemotherapy warrants further projections, for example to a second impact assessment, to determine for how long this may need to be the case. Potentially there could be recommendations for a second assessment layer to the decision trees (e.g., hypothetically after 10 years of MDA). For many countries, reaching the point of a second impact assessment will be reached within the 2020-2030 time period.

The authors recommend co-administered albendazole-ivermectin against T. trichiura, and also community-wide treatment when hookworm is the dominant infection. This differentiation of some of the STH is welcomed, but WHO guidelines are for undifferentiated STH. Can the authors please advise what level of T. trichiura prevalence/intensity would be the recommended ‘trigger’ for co-administration? What level of hookworm prevalence/intensity would be the trigger for risk mitigation activities such as expanding treatments to additional cohorts? Were helminth-specific cutpoints considered in the modelling to arrive at the optimum treatment requirements? If not, is there intent to research this further, with the aim of providing additional advice for implementation considerations? Similarly, can the authors please indicate what aspects of “effective WASH” were modelled – was this access to a latrine, access to a water source, combined interventions, etc?

What is concerning is that the insights from the NTD Modelling Consortium highlight that this new Roadmap goal will require more investment, and more drugs, for STH, but donations for both are likely to decrease. The clarion call from this manuscript is that unless there is careful long-term sustainability and alternative delivery planning, it could well be that the 2030 STH goal will be too ambitious for many settings.

Is the rationale for the Open Letter provided in sufficient detail? Yes

Does the article adequately reference differing views and opinions? Yes

Are all factual statements correct, and are statements and arguments made adequately supported by citations?

Page 9 of 13 Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

Yes

Is the Open Letter written in accessible language? Yes

Where applicable, are recommendations and next steps explained clearly for others to follow? Partly

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: Soil-transmitted helminth and schistosomiasis epidemiology, global health policy, public health policy and implementation.

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Author Response 02 Dec 2019 Luc Edgar Coffeng,

Reviewer 2: Is the authors’ recommendation of 75% community-wide coverage applicable to all settings, or has this extended beyond what is required for the current morbidity control goal?

Answer: As mentioned in the paper, community-wide coverage of 75% is especially important for settings where hookworm is the dominant STH. This is because of the age- prevalence profile of hookworm where adults bear the main burden of infection. In settings where A. lumbricoides or T. trichiura are the dominant species 75% coverage in adults is probably not strictly necessary provided that the coverage in pre-SAC and SAC is sufficiently high, because the age-prevalence profile peaks in children. We have not run simulations to determine the minimum PC coverage of adults required to reach the morbidity target in settings where A. lumbricoides or T. trichiura are the dominant species, given that 75% coverage of preSAC and SAC is achieved. The minimum PC coverage of adults required in these settings will depend on the baseline prevalence and transmission intensity.

Reviewer 2: The authors are kindly requested to elucidate further on the statement (Discussion) that “it is predicted that EOT can be achieved in low-transmission settings where Ascaris lumbricoides or T. trichiura are the dominant parasites by annual treatment of SAC”. If this is the case, but this manuscript has investigated achieving the goal for scenarios above 20%, it seems unlikely that EOT would enter into consideration and thus, preventive chemotherapy would have to continue.

Answer: The conclusion that EOT can be reached in low-transmission settings comes from earlier simulation studies that classified transmission settings using the basic reproductive number R0, rather than baseline prevalence (1). Any given value of R0 does not directly translate into a particular prevalence of infection, because the prevalence also depends on

Page 10 of 13 Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

other model parameters, such as the aggregation parameter k (describes the tendency that a few individuals will be infected by many worms while most individuals will have low- intensity infections) and the density-dependence of female worm fecundity. However, in general, low-prevalence settings will also have low transmission intensities (low R0) values. Therefore, one can predict that EOT can be achieved in low-transmission/low-prevalence settings. Moreover, it is possible that some moderate-prevalence settings also have low- transmission intensity as measured by R0. This needs to be estimated on a case-by-case basis. In these settings, EOT can also be achieved. It is true that in settings where EOT cannot be achieved using current MDA strategies (all high-prevalence settings and some moderate-prevalence settings) and WASH facilities are not improved to the degree that transmission ceases, MDA would have to be continued indefinitely.

Reviewer 2: If neither WASH access nor EOT are feasible, then preventive chemotherapy needs to be sustained. This recommendation for indefinite preventive chemotherapy warrants further projections, for example to a second impact assessment, to determine for how long this may need to be the case. Potentially there could be recommendations for a second assessment layer to the decision trees (e.g., hypothetically after 10 years of MDA).

Answer: We agree with Reviewer 2 that repeated impact assessments are necessary where EOT cannot be achieved by MDA and/or WASH measures.

Reviewer 2: The authors recommend co-administered albendazole-ivermectin against T. trichiura, and also community-wide treatment when hookworm is the dominant infection. This differentiation of some of the STH is welcomed, but WHO guidelines are for undifferentiated STH. Can the authors please advise what level of T. trichiura prevalence/intensity would be the recommended ‘trigger’ for co-administration? What level of hookworm prevalence/intensity would be the trigger for risk mitigation activities such as expanding treatments to additional cohorts? Were helminth-specific cutpoints considered in the modelling to arrive at the optimum treatment requirements?

Answer: In the simulation models, cut-off points were investigated for each helminth species separately. We did not investigate settings where several STH species are present simultaneously. However, we would expect that cut-off values for one species would still be valid, regardless for the presence of other STH species. It follows that co-administration of albendazole-ivermectin in the case of T. trichiura or community-wide treatment in the case of hookworm are recommended when the prevalence of either species is above 20%. We have not investigated if intensified PC strategies are necessary where species-specific prevalences are lower.

Reviewer 2: Similarly, can the authors please indicate what aspects of “effective WASH” were modelled – was this access to a latrine, access to a water source, combined interventions, etc?

Answer: The WASH simulation study (2) distinguished two WASH modalities: sanitation, which reduces individuals’ contributions to environmental contamination, for example, access to latrines; and hygiene, which reduces individuals’ exposure to infection, for

Page 11 of 13 Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

example hand washing, access to clean water sources. Both sanitation and hygiene measures in isolation and in combination were investigated. One interesting finding was that the impact of hygiene is determined more by the effectiveness of the intervention than its overall uptake, whereas the impact of sanitation depends more directly on the product of uptake and the effectiveness.

1. Truscott JE, Hollingsworth TD, Brooker SJ, Anderson RM. Can chemotherapy alone eliminate the transmission of soil transmitted helminths? Parasites & vectors. 2014;7:266. 2. Coffeng LE, Vaz Nery S, Gray DJ, Bakker R, de Vlas SJ, Clements ACA. Predicted short and long-term impact of deworming and water, hygiene, and sanitation on transmission of soil-transmitted helminths. PLoS neglected tropical diseases. 2018;12(12):e0006758.

Competing Interests: The author declare no competing interests.

Reviewer Report 05 November 2019 https://doi.org/10.21956/gatesopenres.14216.r28131

© 2019 Imtiaz R. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Rubina Imtiaz Children Without Worms, The Task Force for Global Health, Decatur, GA, USA

This is a very well written Open Letter which clearly presents the background on Soil Transmitted Helminths, their health effects on risk populations, the WHO goals and treatment guidelines. The authors then state the proposed new, 2030 goals for STH "morbidity control" defined by <2% prevalence of medium- to high-intensity infections in pre-school age and school age children.

Two mathematical models were used to assess the feasibility and relative efficiency of WHO- proposed end goals, in light of current diagnostic and program-decisional thresholds. Good news is that these models are largely comparable. Key outcomes and recommendations from the models are nicely summarized in Table 1. Both models predict that the new target is feasible in moderate prevalence settings (20-50% in SAC) in the absence of systematic non-access to PC. For highly endemic settings (prevalence> 50%), they recommend semi-annual PC including adults, and dual PC for Trichuris. Shortcomings of the current diagnostic of choice (Kato Katz) are well discussed with options for newer, more sensitive diagnostics open (e.g. PCR methods in low prevalence settings).

It’s very good to see the recommendation to have the flexibility in current indicator thresholds by WHO to accommodate future, more sensitive tests (link indicator threshold to standardized diagnostic procedure or to the true infection prevalence from which specific diagnostic

Page 12 of 13 Gates Open Research Gates Open Research 2019, 3:1632 Last updated: 27 JUL 2021

procedures and sampling designs can be derived).

An additional consideration that may strengthen the model and recommendations for advanced programs, is to consider modeling for specific worms within the STH group: STH comprises 3 discrete parasites, each with its own different drug responsiveness and life cycle (e.g. hookworms have a shorter environmental life cycle, so more prone to "elimination" if adult worm reservoir is effectively reduced through sustained, high-level treatment cycles). So, the baseline data on species-specific burden of STH in an area, and appropriate treatment would be ideal.

In the absence of large scale quality data on STH morbidity, this article addresses the current issues very competently with some feasible recommended actions.

Is the rationale for the Open Letter provided in sufficient detail? Yes

Does the article adequately reference differing views and opinions? Yes

Are all factual statements correct, and are statements and arguments made adequately supported by citations? Yes

Is the Open Letter written in accessible language? Yes

Where applicable, are recommendations and next steps explained clearly for others to follow? Yes

Competing Interests: No competing interests were disclosed.

Reviewer Expertise: infectious disease epidemiology, public health program implementation and global health policy

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

Page 13 of 13